578 ERNST J. SCHREINER 



VARIETAL BLENDS 



The evidence from crop plants indicates that increased population 

 buffering to enhance environmental adaptability of mass-produced, rust- 

 resistant hybrid or synthetic varieties of forest trees could be obtained 

 through seed blends. 



Probst (1957) reported performance studies on 3 varieties of soy- 

 beans grown separately and on 15 different blends of these 3 varieties. 

 The blends were not superior in yield over the highest yielding variety, 

 but there was a marked variety x season interaction. In this respect, 

 blending had a stabilizing effect on yield and appeared to be of importance 

 in approaching maximum yield each year. 



Browning (1957) grew 2 varieties of oats (one susceptible and the 

 other resistant to race 7 of the stem rust fungus) in pure stands and in 

 50-50 mixtures subjected to an induced epiphytotic of race 7. In 2 

 successive years, the yield of the mixture was above the average of the 

 yields of the component varieties grown in pure stands, and much less 

 rust developed on the susceptible variety in the field blend. 



Patterson et at. (1963) compared 6 varieties of oats in pure stands 

 and in equal blends of the 6 in all combinations of 2 for 4 years. 

 Improvement from breeding with parental varieties similar to those com- 

 pared in the blends greatly exceeded improvement from the blends; they 

 concluded that blends may provide a useful interim improvement. 



Pfahler (1964) studied the fitness of 6 homogeneous varieties of 

 Avena and a heterogeneous composite of all 6 varieties. Differences 

 between the varieties were significant at the 1 percent level; the fit- 

 ness of the composite exceeded the mean of the 6 varieties. 



MULT I LINEAL VARIETIES 



Pure-line breeding has been the almost exclusive procedure for the 

 improvement of self-pollinating crop plants. Rust-resistant varieties 

 (lines) of wheat and oats have been developed, but the average maximum 

 duration of effective protection by a particular type of resistance in 

 wheat has been about 15 years (Borlaug, 1966); in oats, Sprague (1967) 

 estimates that the average useful life of a new rust-resistant variety 

 has been about 5 years. 



Jensen (1952) discussed the possibilities for producing a satis- 

 factory multiline variety of oats through a seed blend of pure lines, 

 because such a multiline variety could be expected to possess a longer 

 varietal life, greater stability of production, broader adaptation to 

 environment, and greater protection against disease. The basic assump- 

 tion in the design of a multiline variety was that the chance for maxi- 

 mum production is subordinated to the principle of stability; and that 

 this feature of stability is always pointed toward a future risk 

 situation, primarily the risk of a disease epidemic. 



Borlaug (1953) and Borlaug and Gibler (1953) proposed the production 

 of a "composite variety" of wheat by a modification of conventional back- 

 cross methods. The "backcross lines" would be developed by crossing a 

 commercial variety to a number of varieties having different types of 

 resistance. The several lines would be multiplied separately and mixed 



